End of training

all_results.json

@@ -1,4 +1,5 @@
 {
+    "epoch": 4.0,
     "predict_f1": 0.906,
     "predict_gen_len": 30.77916004940886,
     "predict_loss": 1.6894419193267822,
@@ -11,5 +12,9 @@
     "predict_runtime": 3437.6578,
     "predict_samples": 11334,
     "predict_samples_per_second": 3.297,
-    "predict_steps_per_second": 0.412
+    "predict_steps_per_second": 0.412,
+    "train_loss": 1.8367231689453125,
+    "train_runtime": 11495.8338,
+    "train_samples_per_second": 6.959,
+    "train_steps_per_second": 0.217
 }

train_results.json

@@ -0,0 +1,7 @@
+{
+    "epoch": 4.0,
+    "train_loss": 1.8367231689453125,
+    "train_runtime": 11495.8338,
+    "train_samples_per_second": 6.959,
+    "train_steps_per_second": 0.217
+}

trainer_state.json

@@ -0,0 +1,124 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8,
+      "learning_rate": 1.6000000000000003e-05,
+      "loss": 2.0887,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_f1": 0.9023,
+      "eval_gen_len": 31.24818181818182,
+      "eval_loss": 1.7361507415771484,
+      "eval_precision": 0.9035,
+      "eval_recall": 0.9015,
+      "eval_rouge1": 0.4326,
+      "eval_rouge2": 0.1871,
+      "eval_rougeL": 0.3375,
+      "eval_rougeLsum": 0.3373,
+      "eval_runtime": 386.0456,
+      "eval_samples_per_second": 2.849,
+      "eval_steps_per_second": 0.357,
+      "step": 625
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 1.2e-05,
+      "loss": 1.8362,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_f1": 0.905,
+      "eval_gen_len": 30.303636363636365,
+      "eval_loss": 1.6843606233596802,
+      "eval_precision": 0.9071,
+      "eval_recall": 0.9032,
+      "eval_rouge1": 0.4466,
+      "eval_rouge2": 0.1942,
+      "eval_rougeL": 0.3511,
+      "eval_rougeLsum": 0.3507,
+      "eval_runtime": 351.8932,
+      "eval_samples_per_second": 3.126,
+      "eval_steps_per_second": 0.392,
+      "step": 1250
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 8.000000000000001e-06,
+      "loss": 1.7784,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_f1": 0.9056,
+      "eval_gen_len": 30.79909090909091,
+      "eval_loss": 1.6666187047958374,
+      "eval_precision": 0.907,
+      "eval_recall": 0.9045,
+      "eval_rouge1": 0.451,
+      "eval_rouge2": 0.1992,
+      "eval_rougeL": 0.3554,
+      "eval_rougeLsum": 0.3551,
+      "eval_runtime": 352.5825,
+      "eval_samples_per_second": 3.12,
+      "eval_steps_per_second": 0.391,
+      "step": 1875
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 1.7543,
+      "step": 2000
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0,
+      "loss": 1.7261,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "eval_f1": 0.9064,
+      "eval_gen_len": 30.85090909090909,
+      "eval_loss": 1.6605653762817383,
+      "eval_precision": 0.9078,
+      "eval_recall": 0.9053,
+      "eval_rouge1": 0.4557,
+      "eval_rouge2": 0.2019,
+      "eval_rougeL": 0.3603,
+      "eval_rougeLsum": 0.3597,
+      "eval_runtime": 359.5712,
+      "eval_samples_per_second": 3.059,
+      "eval_steps_per_second": 0.384,
+      "step": 2500
+    },
+    {
+      "epoch": 4.0,
+      "step": 2500,
+      "total_flos": 1.9957763220524237e+17,
+      "train_loss": 1.8367231689453125,
+      "train_runtime": 11495.8338,
+      "train_samples_per_second": 6.959,
+      "train_steps_per_second": 0.217
+    }
+  ],
+  "logging_steps": 500,
+  "max_steps": 2500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 4,
+  "save_steps": 500,
+  "total_flos": 1.9957763220524237e+17,
+  "train_batch_size": 8,
+  "trial_name": null,
+  "trial_params": null
+}